Institute of research into the fundamental laws of the Universe

Gamma spectrometry analysis is a classic technique used for the identification and quantification of radionuclides in a radioactive source in strategic applications such as border surveillance, decommissioning, etc. Current analytical methods are based on spectral unmixing, using a library of characteristic spectra (spectral signatures), for each radionuclide to be identified. The case of field measurements on portable devices poses significant algorithmic challenges. On the one hand, the spectral signatures are fixed, severely limiting the robustness of the identification algorithms to the variability of the conditions of field measurements; the phenomena of attenuation or diffusion around a radioactive source lead to strong distortions of the measured spectra. On the other hand, both the small amount of data available to characterize the wide variety of measurable signatures in the field, and the use of portable devices require the development of machine learning algorithms. The objective of the thesis is the development of an algorithmic solution combining fast algorithms with a sober machine learning model, requiring little training data, for the joint estimation of activities and spectral signatures of the radionuclides to be identified. This solution will be developed so as to be implemented on a low-energy on-board digital system (FPGA). The thesis work will also focus on the assessment of the analysis performance balance (detection and identification) and energy sobriety.

With the advent in astrophysics of telescopes embarking various multi/hyperspectral instruments providing complementary measurement modalities, the development of adapted data fusion methods opens the way to a new approach for the reconstruction of images with both high spectral and spatial resolution, a key element for optimal exploitation of the data provided by these instruments. The objective of this doctoral project is the development of data fusion methods that account for all the complexity of real multimodal images: incomplete coverage of the images to be analyzed, complex noises. To this end, we will develop methods inspired by recent methods in machine learning (algorithm unrolling). This approach will allow the construction of hybrid algorithms to take into account the entire complexity of the measurements and the learning of learned regularizations adapted to the data to be reconstructed, thus allowing a significant gain in the interpretability of the exploitability of the results. The tools developed will be applied and validated with realistic

simulated data from the Athena X-ray telescope as well as real data from the XRISM telescope.

Our way to observe the sky have changed with the emergence of the multi-messenger astronomy. In order to efficiently detect electromagnetic counterparts of gravitational waves coming from the most compact objects, we need to develop new instruments pushing back limitations in sensitivity and angular resolution in the hard X-ray domain for a quick detection with low source confusion.



To follow progress in X-ray optics, new fine-pitch large area space detector systems shall be invented. In a similar way, the exploration of the solar system, whether for space weather or for the intensification of the Martian and Lunar exploration will benefit from miniature X/gamma-ray detectors.

Such a challenge has to be firstly solved at the level of the front-end electronics and its interconnects with sensors.



This thesis consists in the development of a new microelectronics circuit (ASIC) to be interconnected with a semiconductor detector (CdTe or Si). This circuit aims to be a matrix of 32 x 32 pixels with 250 x 250 µm² size each of them embedding a full charge readout system.



Since 2011 our team develops a new concept of hybrid detector called MC2 (Mini CdTe on Chip) which relies on 3D technologies such as WDoD (wireless die on die) and able to withstand the space environment. The ambition is to realize large focal plane with innovative spectro-imaging performance with time and polarimetry resolution for next X and gamma ray missions.



The targeted microelectronics technology is the XFAB 180 nm, particularly attractive for space missions thanks to its durable commercial and affordable availability. It constitutes a believable alternative to AMS 0.35 um widely used for space missions such as SVOM (ECLAIRs) and Solar Orbiter (STIX). On top of that, this technology has been migrated recently from Malaysia to France, an interesting perspective for the design optimization in the coming years.



Thanks to two PhD thesis, our group has designed two generations of circuits in this technology, which have shown promising results for the integration of self-triggered spectroscopic chain, with ultra-low noise and low power pixels of 250 x 250 µm². From a system point of view, these circuits have also shown the necessity to develop new blocks and to revise in depth front-end blocks to reach our ultimate low noise performances.



The objective of this thesis is to bring new solutions for a 2-side buttable large array circuit with 250 um pixel pitch, with an optimized modular architecture for its readout.

In current and future high-energy physics experiments: LHC large detectors upgrades and experiments on future colliders, the granularity of particle detectors continues to increase and the use of multi-channel submicron integrated circuits has become a standard. Increasingly, signals from the detectors will have to be digitized by readout circuits and carried away from the experiment by ultra-fast links. The development of new fast and low-power analog-to-digital converters (ADC) operating in often extreme environments, in particular in terms of radiation, is a challenge. Located inside detectors, ADCs are subject to environmental constraints: temperature variations, aging and radiation. The trend has been to try to make the responses of these circuits as stable and independent as possible of variations in environmental parameters (T°, dose and time) and technological parameters (process variation and mismatch). Another approach is to establish precise calibration tables that can be "downloaded" into the generic ASIC as conditions change or generated automatically by the ASIC.

Axions were introduced as the most promising solution in explaining the absence of Charge-Parity symmetry violation in the strong interaction. These neutral, very light particles, interact so weakly with ordinary matter that they could contribute to the Dark Matter. Axion search techniques rely on their interaction with photons. Helioscopes search for axions produced in the solar core by the conversion of plasma photons into axions giving rise to a solar axion flux at the Earth surface, with energy spectrum at the region 1-10 keV.

The International Axion Observatory (IAXO) will achieve a signal-to-background ratio of about 4-5 orders of magnitude better than most sensitive experiments today. BabyIAXO, an intermediate experimental stage of IAXO, will be hosted at DESY (Germany). BabyIAXO is conceived to test all IAXO subsystems (magnet, optics and detectors) at a relevant scale for the final system and thus serve as prototype for IAXO, but at the same time as a fully-fledged helioscope with relevant physics reach in itself, and with potential for discovery. IAXO and BabyIAXO will be equipped with X-ray optics coupled to low background X-ray detectors. The required levels of background are extremely challenging, a factor 10 better than current levels.

The PhD will work on the X-ray detector development in particular of the new generation of Micromegas detectors. The development will be focused on the optimization of the background level by a multi-approach strategy coming from ground measurements, screening campaigns of components of the detector, underground measurements, background models, in-situ background measurements as well as refinement of rejection algorithms. Physics analysis of BabyIAXO data is expected in the last year of the PhD.

Future experiments on linear colliders (e+e-) with low hadronic background require improvements in the spatial resolution of pixel vertex detectors to the micron range, in order to determine precisely the primary and secondary vertices for particles with a high transverse momentum. This kind of detector is set closest to the interaction point. This will provide the opportunity to make precision lifetime measurements of short-lived charged particles. We need to develop pixels arrays with a pixel dimension below the micron squared. The proposed technologies (DOTPIX: Quantum Dot Pixels) should give a significant advance in particle tracking and vertexing. Although the principle of these new devices has been already been studied in IRFU (see reference), this doctoral work should focus on the study of real devices which should then be fabricated using nanotechnologies in collaboration with other Institutes. This should require the use of simulation codes and the fabrication of test structures. Applications outside basics physics are X ray imaging and optimum resolution sensors for visible light holographic cameras.

The ENUBET (Enhanced NeUtrino BEams from kaon Tagging) project aims at building a monitored neutrino beam to reduce the uncertainty on the neutrino flux and cross section below 1%. Given the high rate of events expected in ENUBET, detector time resolution is a critical parameter for clean reconstruction of the events and strong reduction of the mixing of different events due to pile-up.

Furthermore, sub-ns sampling in the far detector would allow one-to-one correlation between positrons tagged in the beamline and neutrinos tagged in the far detector, transforming ENUBET in the first “tagged neutrino beam”.



PIMENT is an ANR-funded R&D project to develop novel detector instrumentation based on the PICOSEC-Micromegas concept and demonstrate the impact of such detectors to New Physics searches by utilizing them to flavor and time tag neutrino beams. Possible exploitation of the PICOSEC-Micromegas technology will be investigated for both the ENUBET tagger and the neutrino detectors.



In the concept of the proposed thesis, the successful candidate will: a) perform physics case studies on the impact of the use of the PICOSEC Micromegas technology in ENUBET for different scenarios: PICOSEC Micromegas as T0 layers, PICOSEC Micromegas embedded in the Electromagnetic Callorimeter, instrumentation of the hadron dump, time tagging at the Liquid Argon Callorimeter; b) participate on the development of modular, multi-channel prototypes (~100 of pads) equipped with novel, carbon-based photocathodes with sufficient photo-electron yield, adapted to the specific needs of each scenario; c) study the timing performance of the prototypes at the IRAMIS fs UV laser at CEA/Saclay and at CERN particle beams; d) evaluate the performance of the prototypes using novel electronics, based on on the SAMPIC readout circuit.